big-data/hadoop
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
21,628 Commits 2 Branches 2 Tags 264 MiB
fc05b0e70e
Commit Graph

1638 Commits

Author SHA1 Message Date
Yufei Gu
e89941fdbb YARN-9041. Performance Optimization of method FSPreemptionThread#identifyContainersToPreempt. Contributed by Wanqiang Ji. 2018-12-04 10:08:45 -08:00
Haibo Chen
a41b648e98 YARN-8992. Fair scheduler can delete a dynamic queue while an application attempt is being added to the queue. (Contributed by Wilfred Spiegelenburg) 2018-11-20 14:12:37 -08:00
Sunil G
5fb14e0635 YARN-8953. [CSI] CSI driver adaptor module support in NodeManager. Contributed by Weiwei Yang. 2018-11-20 08:02:11 +05:30
Weiwei Yang
d027a24f03 YARN-8833. Avoid potential integer overflow when computing fair shares. Contributed by liyakun. 2018-11-18 23:18:26 +08:00
Weiwei Yang
c741109522 YARN-8987. Usability improvements node-attributes CLI. Contributed by Bibin A Chundatt. 2018-11-12 18:18:23 +08:00
Sunil G
4e7284443e YARN-8902. [CSI] Add volume manager that manages CSI volume lifecycle. Contributed by Weiwei Yang. 2018-11-12 11:57:02 +05:30
Weiwei Yang
c96cbe8659 YARN-8977. Remove unnecessary type casting when calling AbstractYarnScheduler#getSchedulerNode. Contributed by Wanqiang Ji. 2018-11-07 22:42:22 +08:00
Akira Ajisaka
951c98f890
YARN-8233. NPE in CapacityScheduler#tryCommit when handling allocate/reserve proposal whose allocatedOrReservedContainer is null. Contributed by Tao Yang. 2018-11-07 11:17:35 +09:00
Jason Lowe
ab6aa4c726 YARN-8865. RMStateStore contains large number of expired RMDelegationToken. Contributed by Wilfred Spiegelenburg 2018-11-06 08:40:59 -06:00
Jonathan Hung
2ab611d48b YARN-7225. Add queue and partition info to RM audit log. Contributed by Eric Payne 2018-11-01 14:22:00 -07:00
Weiwei Yang
1d90a0dd23 YARN-8944. TestContainerAllocation.testUserLimitAllocationMultipleContainers failure after YARN-8896. Contributed by Wilfred Spiegelenburg. 2018-10-29 11:53:10 +08:00
Jason Lowe
93fb3b4b9c YARN-8904. TestRMDelegationTokens can fail in testRMDTMasterKeyStateOnRollingMasterKey. Contributed by Wilfred Spiegelenburg 2018-10-23 12:49:15 -05:00
Haibo Chen
285d2c0753 YARN-8449. RM HA for AM web server HTTPS Support. (Contributed by Robert Kanter) 2018-10-18 21:24:36 -07:00
Weiwei Yang
13cc0f50ea YARN-8907. Fix incorrect logging message in TestCapacityScheduler. Contributed by Zhankun Tang. 2018-10-19 09:55:59 +08:00
Sunil G
5085e5fa9e YARN-8759. Copy of resource-types.xml is not deleted if test fails, causes other test failures. Contributed by Antal Bálint Steinbach. 2018-10-17 16:05:08 +05:30
Haibo Chen
84e22a6af4 YARN-8842. Expose metrics for custom resource types in QueueMetrics. (Contributed by Szilard Nemeth) 2018-10-16 14:14:30 -07:00
Haibo Chen
c2288ac45b YARN-8448. AM HTTPS Support for AM communication with RMWeb proxy. (Contributed by Robert Kanter) 2018-10-16 13:36:26 -07:00
Weiwei Yang
edce866489 YARN-8858. CapacityScheduler should respect maximum node resource when per-queue maximum-allocation is being used. Contributed by Wangda Tan. 2018-10-10 09:32:27 +08:00
Haibo Chen
7c13872cbb YARN-8659. RMWebServices returns only RUNNING apps when filtered with queue. (Contributed by Szilard Nemeth) 2018-10-08 09:40:37 -07:00
Jason Lowe
9bb2801e8c YARN-4254. ApplicationAttempt stuck for ever due to UnknownHostException. Contributed by Bibin A Chundatt 2018-10-05 15:52:46 -05:00
Haibo Chen
c968365650 YARN-8644. Improve unit test for RMAppImpl.FinalTransition. (Contributed by Szilard Nemeth) 2018-10-05 09:32:30 -07:00
Haibo Chen
e60b797c88 YARN-8750. Refactor TestQueueMetrics. (Contributed by Szilard Nemeth) 2018-10-04 13:00:31 -07:00
Haibo Chen
b6d5d84e07 YARN-8732. Add unit tests of min/max allocation for custom resource types in FairScheduler. (Contributed by Szilard Nemeth) 2018-10-04 12:47:31 -07:00
Haibo Chen
d0ee6fbe28 YARN-8621. Add test coverage of custom Resource Types for the apps/<appId> REST API endpoint. (Contributed by Szilard Nemeth) 2018-10-01 14:46:42 -07:00
Weiwei Yang
fd6be5898a YARN-8468. Enable the use of queue based maximum container allocation limit and implement it in FairScheduler. Contributed by Antal Bálint Steinbach. 2018-09-29 17:47:12 +08:00
Eric E Payne
8598b498bc YARN-8774. Memory leak when CapacityScheduler allocates from reserved container with non-default label. Contributed by Tao Yang. 2018-09-28 15:32:07 +00:00
Jason Lowe
6b988d821e YARN-8804. resourceLimits may be wrongly calculated when leaf-queue is blocked in cluster with 3+ level queues. Contributed by Tao Yang 2018-09-26 14:43:00 -07:00
Rohith Sharma K S
e5287a4fe0 YARN-8824. App Nodelabel missed after RM restart for finished apps. Contributed by Bibin A Chundatt. 2018-09-26 12:30:26 +05:30
Akira Ajisaka
44edcdfd6a
YARN-8745. Misplaced the TestRMWebServicesFairScheduler.java file. Contributed by Y. SREENIVASULU REDDY. 2018-09-26 10:09:11 +09:00
Rohith Sharma K S
50bc7746d7 YARN-8815. RM fails to recover finished unmanaged AM. Contributed by Bibin A Chundatt. 2018-09-25 11:31:14 +05:30
Weiwei Yang
0712537e79 YARN-8771. CapacityScheduler fails to unreserve when cluster resource contains empty resource type. Contributed by Tao Yang. 2018-09-19 19:31:07 +08:00
Weiwei Yang
f1a893fdbc YARN-8720. CapacityScheduler does not enforce max resource allocation check at queue level. Contributed by Tarun Parimi. 2018-09-14 16:33:51 +08:00
Sunil G
5e64e62dee YARN-8740. Clear node attribute path after each test run. Contributed by Bibin A Chundatt. 2018-09-12 16:01:01 +05:30
bibinchundatt
c44088ac19 YARN-8739. Fix jenkins issues for Node Attributes branch. Contributed by Sunil Govindan. 2018-09-12 16:01:01 +05:30
Weiwei Yang
52194351e7 YARN-8721. Relax NE node-attribute check when attribute doesn't exist on a node. Contributed by Sunil Govindan. 2018-09-12 16:01:01 +05:30
Naganarasimha
eb08543c7a YARN-8103. Add CLI interface to query node attributes. Contributed by Bibin A Chundatt. 2018-09-12 16:01:01 +05:30
bibinchundatt
8cf6a9a2bd YARN-7892. Revisit NodeAttribute class structure. Contributed by Naganarasimha G R. 2018-09-12 16:01:00 +05:30
Naganarasimha
5dc7d6e0f3 YARN-8104. Add API to fetch node to attribute mapping. Contributed by Bibin A Chundatt. 2018-09-12 16:01:00 +05:30
Naganarasimha
0a01b1350d YARN-8100. Support API interface to query cluster attributes and attribute to nodes. Contributed by Bibin A Chundatt. 2018-09-12 16:01:00 +05:30
Sunil G
b9890d1f66 YARN-7875. Node Attribute store for storing and recovering attributes. Contributed by Bibin A Chundatt. 2018-09-12 16:01:00 +05:30
bibinchundatt
a6590c1f1f YARN-8117. Fix TestRMWebServicesNodes test failure. Contributed by Bibin A Chundatt. 2018-09-12 16:01:00 +05:30
bibinchundatt
901e85238d YARN-8033. CLI Integration with NodeAttributesManagerImpl. Contributed by Naganarasimha G R. 2018-09-12 16:01:00 +05:30
Sunil G
89b3ebd11e YARN-8092. Expose Node Attributes info via RM nodes REST API. Contributed by Weiwei Yang. 2018-09-12 16:01:00 +05:30
Sunil G
6f4bc49c6d YARN-7988. Refactor FSNodeLabelStore code for Node Attributes store support. Contributed by Bibin A Chundatt. 2018-09-12 16:01:00 +05:30
Naganarasimha
3b3b6efe21 YARN-7871. Node attributes reporting from NM to RM. Contributed by Weiwei Yang. 2018-09-12 16:01:00 +05:30
Naganarasimha
86d024ef2a YARN-7965. NodeAttributeManager add/get API is not working properly. Contributed by Weiwei Yang. 2018-09-12 16:01:00 +05:30
Eric E Payne
987d8191ad YARN-8709: CS preemption monitor always fails since one under-served queue was deleted. Contributed by Tao Yang. 2018-09-10 19:55:20 +00:00
Sunil G
1ac01444a2 YARN-8015. Support all types of placement constraint support for Capacity Scheduler. Contributed by Weiwei Yang. 2018-08-23 10:05:43 +05:30
Weiwei Yang
9c3fc3ef28 YARN-7494. Add muti-node lookup mechanism and pluggable nodes sorting policies to optimize placement decision. Contributed by Sunil Govindan. 2018-08-21 22:42:28 +08:00
Rohith Sharma K S
4aacbfff60 YARN-8679. [ATSv2] If HBase cluster is down for long time, high chances that NM ContainerManager dispatcher get blocked. Contributed by Wangda Tan. 2018-08-18 10:26:55 +05:30
Wei-Chiu Chuang
bdd0e01ed9 HADOOP-14212. Addendum patch: Expose SecurityEnabled boolean field in JMX for other services besides NameNode. Contributed by Adam Antal. 2018-08-14 18:25:38 -07:00
Weiwei Yang
0a71bf1452 YARN-8575. Avoid committing allocation proposal to unavailable nodes in async scheduling. Contributed by Tao Yang. 2018-08-10 14:37:45 +08:00
Weiwei Yang
08d5060605 YARN-8521. NPE in AllocationTagsManager when a container is removed more than once. Contributed by Weiwei Yang. 2018-08-10 08:32:02 +08:00
Robert Kanter
b2517dd66b YARN-4946. RM should not consider an application as COMPLETED when log aggregation is not in a terminal state (snemeth via rkanter) 2018-08-09 14:58:04 -07:00
Weiwei Yang
d352f167eb YARN-8559. Expose mutable-conf scheduler's configuration in RM /scheduler-conf endpoint. Contributed by Weiwei Yang. 2018-08-09 23:46:53 +08:00
Haibo Chen
889df6f194 YARN-7948. Enable fair scheduler to refresh maximum allocation for multiple resource types. (Szilard Nemeth via Haibo Chen) 2018-08-02 10:10:12 -07:00
Sunil G
12a095a496 YARN-7159. Normalize unit of resource objects in RM to avoid unit conversion in critical path. Contributed by Manikandan R. 2018-08-02 22:29:21 +05:30
Sunil G
735b492556 YARN-8593. Add RM web service endpoint to get user information. Contributed by Akhil PB. 2018-08-02 08:34:09 +05:30
Sunil G
a48a0cc7fd YARN-8606. Opportunistic scheduling does not work post RM failover. Contributed by Bibin A Chundatt. 2018-08-01 12:17:18 +05:30
Eric Yang
c7ebcd76bf YARN-8579. Recover NMToken of previous attempted component data. 
Contributed by Gour Saha
 2018-07-31 18:01:02 -04:00
Haibo Chen
8aa93a575e YARN-8605. TestDominantResourceFairnessPolicy.testModWhileSorting is flaky. (Wilfred Spiegelenburg via Haibo Chen) 2018-07-31 11:32:40 -07:00
Jonathan Hung
3e06a5dcea YARN-7974. Allow updating application tracking url after registration. Contributed by Jonathan Hung 2018-07-30 17:44:18 -07:00
Robert Kanter
2cccf4061c YARN-8517. getContainer and getContainers ResourceManager REST API methods are not documented (snemeth via rkanter) 2018-07-27 14:35:03 -07:00
Eric Yang
f93ecf5c1e YARN-8330. Improved publishing ALLOCATED events to ATS. 
Contributed by Suma Shivaprasad
 2018-07-25 18:49:30 -04:00
Eric E Payne
9485c9aee6 YARN-4606. CapacityScheduler: applications could get starved because computation of #activeUsers considers pending apps. Contributed by Manikandan R 2018-07-25 16:22:04 +00:00
Weiwei Yang
5be9f4a5d0 YARN-8546. Resource leak caused by a reserved container being released more than once under async scheduling. Contributed by Tao Yang. 2018-07-25 17:35:27 +08:00
bibinchundatt
e673dd1d4d YARN-8541. RM startup failure on recovery after user deletion. Contributed by Bibin A Chundatt. 2018-07-24 23:26:59 +05:30
Sunil G
35ce6eb1f5 YARN-7748. TestContainerResizing.testIncreaseContainerUnreservedWhenApplicationCompleted fails due to multiple container fail events. Contributed by Weiwei Yang. 2018-07-24 22:20:17 +05:30
bibinchundatt
ff7c2eda34 YARN-8548. AllocationRespose proto setNMToken initBuilder not done. Contributed by Bilwa S T. 2018-07-24 16:17:20 +05:30
Haibo Chen
8a6bb8409c YARN-6964. Fair scheduler misuses Resources operations. (Daniel Templeton and Szilard Nemeth via Haibo Chen) 2018-07-20 10:46:28 -07:00
Weiwei Yang
cbf2026483 YARN-8528. Final states in ContainerAllocation might be modified externally causing unexpected allocation results. Contributed by Xintong Song. 2018-07-20 22:34:06 +08:00
Haibo Chen
2564884757 YARN-8436. FSParentQueue: Comparison method violates its general contract. (Wilfred Spiegelenburg via Haibo Chen) 2018-07-19 13:22:31 -07:00
Eric Yang
5836e0a46b YARN-8501. Reduce complexity of RMWebServices getApps method. 
Contributed by Szilard Nemeth
 2018-07-19 12:30:38 -04:00
Wangda Tan
a2e49f41a8 YARN-8361. Change App Name Placement Rule to use App Name instead of App Id for configuration. (Zian Chen via wangda) 
Change-Id: I17e5021f8f611a9c5e3bd4b38f25e08585afc6b1
 2018-07-16 10:57:37 -07:00
Wangda Tan
752dcce5f4 YARN-8511. When AM releases a container, RM removes allocation tags before it is released by NM. (Weiwei Yang via wangda) 
Change-Id: I6f9f409f2ef685b405cbff547dea9623bf3322d9
 2018-07-16 10:54:41 -07:00
Eric E Payne
937ef39b3f YARN-8421: when moving app, activeUsers is increased, even though app does not have outstanding request. Contributed by Kyungwan Nam 2018-07-16 16:24:21 +00:00
Robert Kanter
a129e3e74e YARN-7451. Add missing tests to verify the presence of custom resources of RM apps and scheduler webservice endpoints (snemeth via rkanter) 2018-07-06 11:04:00 -07:00
Haibo Chen
9edc74f64a YARN-7556. Fair scheduler configuration should allow resource types in the minResources and maxResources properties. (Daniel Templeton and Szilard Nemeth via Haibo Chen) 2018-07-06 11:03:48 -07:00
Anu Engineer
39ad98903a Revert "Merge branch 'trunk' of https://git-wip-us.apache.org/repos/asf/hadoop into trunk" 
This reverts commit c163d1797a, reversing
changes made to 0d9804dcef.
 2018-07-05 12:22:18 -07:00
Robert Kanter
99febe7fd5 YARN-7451. Add missing tests to verify the presence of custom resources of RM apps and scheduler webservice endpoints (snemeth via rkanter) 2018-07-05 10:54:19 -07:00
Haibo Chen
1726247024 YARN-7556. Fair scheduler configuration should allow resource types in the minResources and maxResources properties. (Daniel Templeton and Szilard Nemeth via Haibo Chen) 2018-07-05 10:42:39 -07:00
Junping Du
d9ba6f3656 yarn.resourcemanager.fail-fast is used inconsistently. Contributed by Yuanbo Liu. 2018-07-03 14:46:44 +08:00
Sunil G
291194302c YARN-8379. Improve balancing resources in already satisfied queues by using Capacity Scheduler preemption. Contributed by Zian Chen. 2018-06-28 10:23:31 -07:00
Eric Yang
384764cdea YARN-8409. Fixed NPE in ActiveStandbyElectorBasedElectorService. 
Contributed by Chandni Singh
 2018-06-28 12:39:49 -04:00
Inigo Goiri
99948565cb YARN-8412. Move ResourceRequest.clone logic everywhere into a proper API. Contributed by Botong Huang. 2018-06-21 18:24:10 -07:00
Akira Ajisaka
652bcbb3e4
YARN-8363. Upgrade commons-lang version to 3.7 in hadoop-yarn-project. Contributed by Takanobu Asanuma. 2018-06-13 00:35:37 +09:00
Rohith Sharma K S
2df73dace0 YARN-8405. RM zk-state-store.parent-path ACLs has been changed since HADOOP-14773. Contributed by Íñigo Goiri. 2018-06-12 17:13:09 +05:30
Robert Kanter
0cd145a443 YARN-4677. RMNodeResourceUpdateEvent update from scheduler can lead to race condition (wilfreds and gphillips via rkanter) 2018-06-04 15:32:03 -07:00
Haibo Chen
ba12f87dcb YARN-8390. Fix API incompatible changes in FairScheduler's AllocationFileLoaderService. (Gergo Repas via Haibo Chen) 2018-06-04 09:15:23 -07:00
Wangda Tan
931f78718f YARN-7962. Race Condition When Stopping DelegationTokenRenewer causes RM crash during failover. (BELUGA BEHR via wangda) 
Change-Id: I617e2645f60a57080058ad5f06af860fb3f682c8
 2018-06-01 14:49:18 -07:00
Weiwei Yang
6468071f13 YARN-8367. Fix NPE in SingleConstraintAppPlacementAllocator when placement constraint in SchedulingRequest is null. Contributed by Weiwei Yang. 2018-05-31 20:46:39 +08:00
Sunil G
7c343669ba YARN-4781. Support intra-queue preemption for fairness ordering policy. Contributed by Eric Payne. 2018-05-28 16:32:53 +05:30
Wangda Tan
f24c842d52 YARN-8213. Add Capacity Scheduler performance metrics. (Weiwei Yang via wangda) 
Change-Id: Ieea6f3eeb83c90cd74233fea896f0fcd0f325d5f
 2018-05-25 21:53:20 -07:00
Eric E Payne
8d5509c681 YARN-8292: Fix the dominant resource preemption cannot happen when some of the resource vector becomes negative. Contributed by Wangda Tan. 2018-05-25 16:06:09 +00:00
Haibo Chen
86bc6425d4 YARN-8191. Fair scheduler: queue deletion without RM restart. (Gergo Repas via Haibo Chen) 2018-05-24 17:12:34 -07:00
Wangda Tan
bd15d2396e YARN-8290. SystemMetricsPublisher.appACLsUpdated should be invoked after application information is published to ATS to avoid "User is not set in the application report" Exception. (Eric Yang via wangda) 
Change-Id: I0ac6ddd19740d1aa7dd07111cd11af71ddc2fcaf
 2018-05-22 13:33:33 -07:00
Eric E Payne
0b4c44bdee YARN-8179: Preemption does not happen due to natural_termination_factor when DRF is used. Contributed by Kyungwan Nam. 2018-05-21 20:14:58 +00:00
Haibo Chen
f48fec83d0 YARN-8248. Job hangs when a job requests a resource that its queue does not have. (Szilard Nemeth via Haibo Chen) 2018-05-21 08:10:41 -07:00
Arun Suresh
3159bffce2 YARN-7900. [AMRMProxy] AMRMClientRelayer for stateful FederationInterceptor. (Botong Huang via asuresh) 2018-05-17 20:00:52 -07:00
Haibo Chen
1f10a36021 YARN-8268. Fair scheduler: reservable queue is configured both as parent and leaf queue. (Gergo Repas via Haibo Chen) 2018-05-11 11:34:00 -07:00